André Eugênio Lazzaretti

Measurements of lightning discharges in overhead distribution feeders

This paper presents the development of an automatic system designed to measure fast electromagnetic transients on energized distribution network on both, medium and low voltage circuits. The four developed systems were installed on class 15 kV feeders. The feeders were chosen according to the flash lightning density levels of the region. Besides the constructive details of the develop system, the paper shows the main results achieved.

Classification of Events in Distribution Networks using Autonomous Neural Models

This paper presents a method for automatic clas- sification of faults and events related to quality of service in electricity distribution networks. The method consists in preprocessing event oscillographies using the wavelet transform and then classifying them using autonomous neural models. In the preprocessing stage, the energy present in each sub-band of the wavelet domain is computed in order to compose input feature vectors for the classification stage. The classifiers investigated are based in Multi-Layer Perceptron (MLP) feed-forward artificial neural networks and Support Vector Machines (SVM), which automatically promote input selection and structure complexity control simultaneously. Experiments using simulated data show promising results for the proposed application.

An Accurate Approach for Automatic Segmentation of Power Distribution Voltage Waveforms

This paper addresses one of the fundamental steps in automatic waveform analysis: transient segmentation. We present a new approach which incorporates the advantages of a multilevel wavelet decomposition and the representation of the support vector data description. Real data from a monitoring system developed for lightning overvoltage detection in overhead distribution power lines was used for comparison and validation of segmentation performance. The experiments involve the proposed segmentation approach and usual segmentation methods, such as Kalman filtering, autoregressive models, and standard discrete wavelet transform. The results show that the proposed segmentation method based on DWT+SVDD yields better overall accuracy for transient segmentation when compared to currently used methods, demonstrating the potential for applications in oscillographic recorders for smart distribution networks, where identification, characterization, and mitigation of events are critical for network operation and maintenance.

A new approach for event classification and novelty detection in power distribution networks

This paper presents a new approach for automatic oscillography classification in distribution networks, including the detection of patterns not initially presented to the classifier during training, which are defined as novelties. We performed experiments with coupled novelty detection and multi-class classification, and also in separate stages, using the following classifiers: Gaussian Mixture Models (GMM), K-means clustering (KM), K-nearest neighbors (KNN), Parzen Windows (PW), Support Vector Data Description (SVDD), and multi-class classification based on Support Vector Machines (SVM). Preliminary results for simulated data in the Alternative Transient Program (ATP) demonstrate the ability of the method to identify new classes of events in a dynamic learning environment. This work was partially supported by COPEL within the Research and Development Program of the Brazilian Electrical Energy Agency (ANEEL).

A study of deep convolutional auto-encoders for anomaly detection in videos

Deep convolutional auto-encoder for anomaly detection in videos.Fusion of low-level (frames) with high-level (appearance and motion features) information.Study of the influence of video complexity in the classification performance.Use of reconstruction errors from convolutional auto-encoder as anomaly scores.Case studies with real-world video clips. The detection of anomalous behaviors in automated video surveillance is a recurrent topic in recent computer vision research. Depending on the application field, anomalies can present different characteristics and challenges. Convolutional Neural Networks have achieved the state-of-the-art performance for object recognition in recent years, since they learn features automatically during the training process. From the anomaly detection perspective, the Convolutional Autoencoder (CAE) is an interesting choice, since it captures the 2D structure in image sequences during the learning process. This work uses a CAE in the anomaly detection context, by applying the reconstruction error of each frame as an anomaly score. By exploring the CAE architecture, we also propose a method for aggregating high-level spatial and temporal features with the input frames and investigate how they affect the CAE performance. An easy-to-use measure of video spatial complexity was devised and correlated with the classification performance of the CAE. The proposed methods were evaluated by means of several experiments with public-domain datasets. The promising results support further research in this area.

An integrated monitoring and data analysis system to evaluate the effects of lightning activity on distribution networks

Overhead power distribution lines have significant power quality variations (voltage transients) during thunderstorms. Lightning strokes, whether direct or indirect, are among the main causes of such power quality disturbances. In order to assist the analysis of lightning effects on power quality indices, we present in this paper a monitoring network and an integrated database analysis, including voltage waveform recorders, local electric field measurement, information from Lightning Location Systems, and utility databases. The idea is to present a general procedure to indicate the correlation between lightning and power quality events (mainly short-circuits) in the monitored distribution network. Also, we present an automatic waveform analysis that allows the automatic segmentation and classification of transients in a waveform and a simulation procedure that assists the correlation process. Finally, we present an initial case study addressing the advantages and relevant correlations that can be obtained from the proposed approach.

An Adaptive Radial Basis Function Kernel for Support Vector Data Description

For one-class classification or novelty detection, the metric of the feature space is essential for a good performance. Typically, it is assumed that the metric of the feature space is relatively isotropic, or flat, indicating that a distance of 1 can be interpreted in a similar way for every location and direction in the feature space. When this is not the case, thresholds on distances that are fitted in one part of the feature space will be suboptimal for other parts. To avoid this, the idea of this paper is to modify the width parameter in the Radial Basis Function (RBF) kernel for the Support Vector Data Description (SVDD) classifier. Although there have been numerous approaches to learn the metric in a feature space for (supervised) classification problems, for one-class classification this is harder, because the metric cannot be optimized to improve a classification performance. Instead, here we propose to consider the local pairwise distances in the training set. The results obtained on both artificial and real datasets demonstrate the ability of the modified RBF kernel to identify local scales in the input data, extracting its general structure and improving the final classification performance for novelty detection problems.

Visualization and analysis of atmospheric electric field local profile - REMOTE-E network

Direct and indirect lightning strokes can cause severe overvoltages in distribution networks and, consequently, failures and power outages, due to insulation problems and equipment damages along the network. In order to assist the analysis of lightning effects in power quality indices, we present in this paper, a detailed analysis of the local electric field, combined with the information from Lightning Location Systems (LLS) and voltage waveforms recorded in the distribution network. The proposed analysis is performed in a monitoring network that was particularly designed for the approach presented in this paper. The monitoring network is composed by four Campbell Field Mill sensors, four waveform recorders and the LLS information. Based on the local electric field profile, it is possible to confirm (or reject) lightning activity in the region under analysis, by comparing the information recorded by these three sources of information. Finally, we present some initial case studies addressing the advantages and main limitations of the proposed approach.

A new approach for event detection in smart distribution oscillograph recorders

This paper presents some new results for a fundamental step in automatic oscillography analysis: transient detection. We performed experiments with usual detection methods, such as the Kalman filter (KF) and autoregressive (AR) models, and we are proposing a new method based on the Discrete Wavelet Transform (DWT) and Support Vector Data Description (SVDD). Data simulated in the Alternative Transient Program (ATP) was generated for comparison and validation of detection performance. The results presented here demonstrate that the proposed detection method based on DWT and SVDD yields better overall performance for the transient detection process when compared to currently used methods such as KF and AR models. These results show the potential for possible embedded applications in automatic oscillographic recorders in smart distribution networks, in which identification, characterization, and mitigation of events is critical for network operation and maintenance.

Power System Fault Data Compression Wavelet Parametric Investigation

This work presents a performance comparison method for wavelet based compression of three-phase signals. Input signals are simulated on ATP software, while the mother wavelets, compression thresholds and transform detail levels are changed. Thousands of compressed output signals are thus obtained, and the compression rate and error are calculated. These performance indicators are analyzed, based on which a group of wavelet compression parameters is obtained.